Skip to content

Latest commit

 

History

History
13 lines (9 loc) · 321 Bytes

README.md

File metadata and controls

13 lines (9 loc) · 321 Bytes

img-embedding

  1. Loads random 1000 images from a folder,
  2. Creates embedding by using a pre-trained vgg11_bn model,
  3. Lowers dimensionality using t-SNE to two dimensions,
  4. Plots the images interactively, one can browse and zoom.

output

Run

  1. Add images to imgs folder
  2. python main.py